Skip to content

Defence is crucial against mistaken identifications

Wrongful convictions and mistaken identifications go hand in hand. The Innocence Project in New York City found that in 101 of 130 post-conviction exonerations, mistaken identifications were the cause of the wrongful convictions. Although DNA evidence has and will exonerate some of the wrongfully convicted, the vigilance of defence counsel is the best defence for the wrongfully accused.

A Criminal Mind by Rosalind Conway

Osgoode Hall Law School recently held a one-day conference on eyewitness identification and testimony. Surprisingly, most of the registrants seemed to be police officers.

I fought an identification case on behalf of a Romany woman charged with seven residential break-and-enters. A dark-complexioned European woman would enter a house whose elderly owners were gardening outside. If confronted, she would say she was looking for her little dog. Jewellery and cash were taken.

There was no physical evidence. The lineup was ridiculous: 10 French Canadians, one East Indian woman and one gypsy, my client. All of the witnesses had at least a 50 per cent chance of selecting the suspect. Surprisingly, she was only positively identified by one eyewitness, one who would surely become a devastatingly convincing witness: a lawyer's wife.

Following the preliminary hearing my client stood charged with five break-ins. Two more witnesses had identified her in the courtroom. And, as the offences were similar, the judge ordered her to stand trial on two more in which there had been no identification.

The trouble was, I learned that there were actually more of these break-ins than she was charged with ? 15 more. Following the preliminary hearing, I disclosed our alibi evidence and sought disclosure of the other break-ins. Refused as not relevant.

I brought a pre-trial disclosure application in Superior Court. The Crown and a lawyer for Ottawa Police fought it, arguing that I was seeking privileged third-party records. Justice Albert Roy ruled that he would review the occurrence reports in chambers and adjourned; the prosecution formed into a scrum.

Not long afterwards I had a long telephone message from the investigator: he now had a reasonable doubt that my client was guilty. The charges were withdrawn.

Few defence counsel are aware that the other 11 people in the Ontario lineups are "foils" ? they are not suspects. A lineup normally contains only one suspect. In some places in North America a lineup has as few as six photographs. The maximum is 20. The photographs are not supposed to be clones of the suspect ? they are to be based on the description of the features received from the eyewitness, the composite drawing, or the security videotape. Ryerson's Professor John Turtle explained that the goal of the lineup is to create some "propitious heterogeneity" (beneficial variety) among the photographs.

The Ontario Police College has a "best practices" guide for police officers which provides a framework for local police forces to establish their own procedures. Regrettably, we have no standard procedures in Canada for composing and administering lineups.

When presenting the photo array, the officer should not indicate that the suspect is present, but rather that someone associated with the case may or may not be present. Separate lineups should be given for each witness and each suspect.

It has become more widely accepted that, generally, the photo spread is not the ideal method for proper identifications. In a draft of an upcoming article, Osgoode Hall Professor Christopher Sherrin notes that sequential lineups are preferable to simultaneous ones because studies have shown that, even when the perpetrator is not present, a relative judgment may be made between the photographs.

The pictures are to be presented in a stack, upside down, and are shown, in random order, sequentially by an officer not involved in the investigation. The viewer is only shown them once, and is shown one photograph at a time. He or she is not told how many images he will be shown.

The Ontario Police College guidelines suggest that, following a selection, the officer state the following: "You have selected a subject. Tell me why you recognize this person."

It became clear at the conference that the police are avoiding the ultimate question: just how certain are you? All statements must be noted, recorded verbatim and signed by the witness. There is no reason not to videotape the procedure, and to back this up with an audiotape. The selected photograph should be shown to the camera.

The alternative, low-tech procedure is to place the photographs in envelopes. The witness draws each one out separately as the officer goes through the stack, and the witness writes comments on the back of each photograph and replaces it in its envelope. Again, the witness does not know when the viewing is coming to an end.

However, the speed at which the identification occurs is obviously relevant. Queen's Professor Roderick Lindsay said that 30 seconds is the maximum for an identification, and that witnesses should never be told to take their time. Instant recognition is important.

Once the procedure is over, there should be no discussion of the choice that the witness has made, and the witness should leave the police station as quickly as possible to avoid contamination of that witness and any other witnesses.

A word about "in-dock" identifications. These are identifications where the client was not selected from the photo array, or where no lineup was administered to the witness. They have little weight, although they are admissible, and it is prudent to advise the jury that there is a very weak link between the confidence level of a witness and the accuracy of that witness, based on the Supreme Court of Canada's 2002 ruling R. v. Hibbert.

Nevertheless, you can still closely cross-examine the identification officer on his training and the practices, guidelines, and policies of his police force.

A positive defence is crucial in these cases. As Justice Louise Arbour wrote in Hibbert, "I think it is important to remember that the danger associated with eyewitness in-court identification is that it is deceptively credible, largely because it is honest and sincere."

John Rosen, defence counsel from Toronto, pointed out that if your case is going to trial, you had better get to work on your defence ? like alibi ? right away.

Rosalind Conway practises criminal law in Ottawa. She can be reached at rosalind.conway@magma.ca

cover image

DIGITAL EDITION

Subscribers get early and easy access to Law Times.

Law Times Poll


It's unknown how widely police in Ontario utilize controversial surveillance techniques that can capture private data from non-targets in criminal investigations. Do you think there should be formal requirements to release this information?
RESULTS ❯